GAN-Tensorflow

Introduction

Implementation of GAN (Generative Adversarial Nets) in Tensorflow

Tasks:

• Simple vanila GAN on MNIST

  • Two Fully-Connected layers for both Discriminator and Generator.
  • Training result in gif:
  

• Simple vanila GAN on MNIST

  • Small ConvNets for both Disctiminator and Generator.
  • Training result in gif:
  

Reference

Generative Adversarial Networks (2014) by Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio.

Understanding the difficulty of training deep feedforward neural networks (2009) by Xavier Glorot, Yoshua Bengio.